3. Intrusion measurement and occupant survival space

In side impact, the occupant rarely benefits from the large front-end crush zone that protects them in a frontal crash. The distance between the striking barrier and the occupant's torso can be measured in tens of millimetres, and the door trim, armrest, and B-pillar are effectively the last line of defence. Intrusion measurement therefore sits at the centre of side impact development: it converts a deformed structure into numbers that can be compared against targets, correlated with dummy injury metrics, and used to decide whether the restraint system or the body structure must change.

This section defines the key intrusion metrics, explains how they are extracted from test data, sets out typical target values and their physical basis, and shows how intrusion correlates with occupant injury risk. A worked example using representative (illustrative, not measured) values demonstrates the full measurement-to-judgement loop.

Defining the measurement reference frame first

All intrusion values in this course use the vehicle coordinate system defined as follows: X positive forward (longitudinal), Y positive to the vehicle's right (lateral), Z positive upward, with the origin at the vehicle datum. Intrusion is reported as displacement of a structure toward the occupant compartment, i.e. along −Y for a left-side impact on the left door, and is expressed as a positive magnitude in millimetres. Any switch of sign convention must be stated explicitly; mixing conventions is one of the most common sources of disagreement between CAE and test teams.

Core intrusion metrics

Four families of metrics cover most side impact programmes:

  • Maximum intrusion (Dmax) — the peak residual or dynamic displacement of a specified point, typically the door inner panel at occupant chest height, the armrest, or the B-pillar at H-point height. Reported separately for dynamic peak and residual (post-test) values.
  • Intrusion velocity (Vint) — the time derivative of intrusion at the same point. This is often more strongly correlated with injury than displacement alone, because a fast-moving door loads the occupant before the restraint can act.
  • Survival space / occupant compartment residual space — the distance between the deformed interior surface and the occupant's seated reference (e.g. the ES-2re dummy's chest or pelvis surface, or the H-point manikin). This is the quantity that actually matters for injury.
  • Intrusion rate into the occupant's zone — the fraction of the original occupant-to-trim gap consumed during the event, useful for comparing different seating positions and door trims.
Report Dmax and Vint at the same physical point and with the same filter (typically CFC 180 or CFC 600 for displacement, CFC 180 for velocity). A displacement compared across different filter classes is not a valid comparison.

Extracting intrusion from test data

In a physical side impact test (for example a moving deformable barrier at 50 km/h, or a pole test), intrusion is measured with string potentiometers, linear potentiometers, or high-speed photogrammetry targets on the door inner panel, armrest, and B-pillar. The raw channel is filtered, zeroed at t = 0 (barrier contact), and then reduced to the metrics above. In CAE, the same points are tracked as nodal displacements in the vehicle frame, and the same post-processing filters are applied so that test and simulation are directly comparable.

A reproducible extraction procedure for one point looks like this:

1. Identify the measurement point P (e.g. door inner panel, chest height, Y = -350 mm from centreline).
2. Zero the displacement channel at barrier contact (t0).
3. Apply CFC 180 filter to displacement; differentiate to obtain V_int, then filter V_int at CFC 180.
4. Record D_max = max(|d(t)|) over the event window.
5. Record V_int,max = max(|v(t)|) over the same window.
6. Record residual d_res = d(t_end) after the event settles.
7. Compute survival space S = S0 - d_res, where S0 is the pre-test occupant-to-trim gap.

The pre-test gap S0 is measured with the dummy seated and the seat at its design position. For a typical mid-size sedan with an ES-2re dummy, S0 at chest height might be around 180–220 mm; at pelvis height, 150–190 mm. These are illustrative ranges, not targets.

Targets and their physical basis

Targets are set from three inputs: regulatory and consumer-test requirements (UN R95, FMVSS 214, Euro NCAP side impact and pole), the OEM's internal injury-metric budget (e.g. rib deflection, V*C, pelvis force), and packaging constraints. A representative target set for a B-pillar/door inner panel at chest height is shown below. Values are illustrative and must be replaced by the programme's own requirements.

MetricIllustrative targetRationale
Dmax (door inner, chest height)≤ 120 mmKeeps residual survival space above the dummy chest contact threshold
Vint,max (same point)≤ 8 m/sLimits loading rate on the chest before restraint inflation
Residual survival space S≥ 75 mmPrevents hard contact between trim and dummy chest
B-pillar intrusion at H-point height≤ 90 mmProtects pelvis and lower torso

Correlation with injury risk

Intrusion does not injure the occupant directly; it injures by reducing the distance over which the restraint can decelerate the torso, and by bringing stiff structure into contact with the body. The correlation is therefore strongest when intrusion is paired with the occupant's loading. Empirically, chest deflection and V*C rise steeply once residual survival space falls below roughly 60–80 mm, and pelvis force rises when B-pillar intrusion at H-point height exceeds about 100 mm. These thresholds are approximate and depend on the dummy, the trim stiffness, and the restraint.

The practical consequence is that intrusion targets should be set together with restraint targets, not in isolation. A door that intrudes 100 mm but is backed by a well-timed side airbag may produce lower injury than a door that intrudes 80 mm with a late-deploying bag. The correlation study is what resolves this trade-off.

Side impact intrusion measurement and survival space correlation Diagram showing the measurement points on a door inner panel, the survival space between trim and dummy, and a correlation curve of chest deflection versus residual survival space. Measurement points and survival space (plan view, left side impact) Vehicle plan view Door inner panel ES-2re torso S (survival space) D_int P1 chest height P2 mid-door P3 H-point height B-pillar Chest deflection vs. residual survival space (illustrative) Residual survival space S (mm) Chest deflection (mm) 0 40 80 120 160 200 0 20 40 60 High-risk zone S < ~70 mm Target S ≥ 75 mm Metric chain: from structure to injury Door / B-pillar displacement D_max, V_int (filtered) Residual survival space S Chest deflection, V*C, pelvis force All values illustrative. Report D_max and V_int at the same point and filter class; state sign convention explicitly. Targets must be set jointly with restraint timing and airbag coverage, not in isolation.

Worked example: from measurement to judgement

Consider a left-side moving deformable barrier test on a mid-size sedan with an ES-2re dummy in the driver seat. The pre-test chest-height gap is S0 = 195 mm. Post-processing of the door inner panel channel (CFC 180) gives Dmax = 132 mm at t = 38 ms, Vint,max = 9.4 m/s at t = 31 ms, and residual dres = 118 mm. The B-pillar at H-point height shows Dmax = 96 mm.

Applying the metrics: residual survival space S = 195 − 118 = 77 mm, which is just above the 75 mm target. However, Dmax = 132 mm exceeds the 120 mm target, and Vint,max = 9.4 m/s exceeds the 8 m/s target. The B-pillar value of 96 mm is within its 90 mm target only marginally. The judgement is therefore not a pass: the door is moving too fast and too far, even though the residual gap looks acceptable. The likely cause is late side airbag deployment relative to door motion, or a door trim that is too compliant in the early phase. The corrective action is to advance airbag timing by 3–5 ms and stiffen the door inner panel locally at chest height, then re-run and re-extract the same channels.

Do not accept a residual survival space that passes while Dmax and Vint fail. Residual space describes the end state; Dmax and Vint describe the loading the occupant actually experiences. A late but large intrusion can still produce high chest deflection because the restraint has already been compressed.

Practical checks before signing off a correlation

  • Confirm the measurement point coordinates match between test and CAE (same Y, Z, same panel location).
  • Confirm the filter class and zeroing time are identical on both sides.
  • Plot intrusion and chest deflection on a common time axis; a phase lag between them indicates restraint timing issues, not structural ones.
  • Check that the dummy seating position and S0 are recorded, since survival space is meaningless without the pre-test gap.
  • Record the sign convention and coordinate frame in the report header so that later readers cannot misinterpret the direction of intrusion.

Used this way, intrusion metrics become a diagnostic instrument rather than a pass/fail checkbox: they tell the engineer whether the structure, the restraint, or the timing is the dominant contributor to occupant injury risk, and they give a quantitative basis for deciding which of the three to change first.

← Previous 📖 Contents Next →